A Unified Model for the Evolution of Galaxies, Active Galactic Nuclei and their central Supermassive Black Holes


Supermassive black holes weighing up to a billion solar masses are found in the central regions of almost all galaxies and have probably played a major role in their formation and evolution. In June of this year a very tight correlation between the mass of a supermassive black hole and the velocity dispersion of its host galaxy has been reported at the summer meeting of the American Astronomical Society. Researchers at the MPA showed immediately that their model for the evolution of galaxies and their central supermassive black holes is in very good agreement with this new correlation (see figure 1). The essence and motivation of this model which was developed in the last year at MPA is described below.

Figure 1: Mass of the supermassive black hole vs. velocity dispersion of its host galaxy for a sample of nearby galaxies. The read and blue data points are from Ferrarese & Merritt and Gebhardt et al.

When supermassive black holes accrete material, a substantial fraction of the binding energy of the infalling matter can be radiated away. The nucleus of the galaxy increases enormously in luminosity and becomes visible as a so-called AGN, or active galactic nucleus .

Much of our knowledge about how supermassive black holes have evolved over cosmic timescales comes from studying the AGN population at different epochs. The space density of bright QSOs which are the dominant class of AGN, exhibits a rapid rise and fall with a conspicuous peak at a redshift of around 2, when the Universe was approximately a fifth of its present age. It is tantalizing that the peak of AGN activity coincides with the epoch at which most of the stars in the Universe were being formed in galaxies.

Another important piece of evidence linking the growth of supermassive black holes to the formation of galaxies was the discovery of a linear correlation between the masses of these black holes and the masses of the galactic bulges that harbour them. At average about 0.3 percent of the mass of the bulge is contained in the central black hole. This suggested that the formation of stars in bulges and the assembly of supermassive black holes at their centres are closely linked. This suggestion is strongly supported by the recently reported much tighter correlation between black hole mass and velocity dispersion of the host galaxy shown in figure 1.

So what is the physical mechanism responsible both for bulge formation and the growth of black holes? Many astronomers now believe that mergers between galaxies are the prime suspect. According to the standard theoretical paradigm for the formation of structure in the Universe, small density inhomogeneities in the dark matter were generated shortly after the Big Bang by quantum fluctuations during a period of accelerated expansion, usually termed inflation. These early inhomogeneities were then gravitationally amplified as the Universe expanded.

Figure2: A spectacular multiple merger imaged by the NICMOS instrument on board the Hubble space telescope.
Eventually, material contained in initially overdense regions of the Universe stopped expanding and began to collapse. The smallest objects formed first in this way, and these later merged together to form larger and larger structures. Galaxies formed in regions where the overdensity of matter was high enough to allow gas to cool, condense and form stars. This scenario is often referred to as the ``bottom-up'' or hierarchical model of structure formation.

Detailed numerical simulations of the merger of two spiral galaxies of equal mass show that the remnant galaxy left after the merger is completed is structurally very similar to observed galactic bulges. During the merging process, tidal forces drive most of the gas in the galaxies to the centre of the remnant, where it is compressed and able to form stars and fuel a central supermassive black hole. Figure 2 shows a spectacular multiple merger in the real Universe imaged by the NICMOS instrument on board the Hubble space telescope.

Many of such merging galaxies have star formation rates of a few hundred solar masses per year and some show evidence of an active galactic nucleus, suggesting that the merger has triggered the accretion of new material onto the central black hole.

Figure 3: V band luminosities of host galaxies vs. B band luminosities of QSOs at two different redshifts.

Last year MPA scientists developed a ``unified'' model for galaxy and AGN evolution that synthesized many of the ideas outlined above. These models track the formation and evolution of structure in the dark matter component of the Universe from standard cosmological initial conditions, as well as the formation and merging of galaxies within the potential wells defined by the dark matter. These models have already met with substantial success in explaining many of the observed properties of galaxies both at the present day and at high redshifts.

The main new ingredient was the hypothesis that supermassive black holes grew in mass during the same major merging events that produced galactic bulges. The MPA researchers demonstrated that the new unified models could not only explain many aspects of galaxy evolution, such as the evolution of the total star formation in galaxies as a function of cosmic epoch, but also the strong rise and fall in the space density of AGN. Perhaps most interestingly, the MPA models made a number of new and observationally testable predictions for the relation between the masses of supermassive black holes and the properties of their galactic hosts, as well as the relation between AGN luminosity and bulge luminosity at different cosmic epochs. Because massive galaxies assemble relatively late in hierarchical cosmogonies, quasars of fixed luminosity were predicted to reside in less massive host galaxies at higher redshifts. In addition, younger galaxies were predicted to contain lower mass black holes in their centres than older galaxies. The first prediction has already been confirmed by a number of recent observations (see figure 3). There has also been some suggestive evidence that black hole mass may indeed correlate with the age of the host galaxy. Future studies will no doubt shed further light on this issue.

For more information see here.

Martin Haehnelt and Guinevere Kauffmann


Comments to: webmaster@mpa-garching.mpg.de